Web App for Visualizing the Distribution of Family Names

As part of the cultural hackathon Coding da Vinci Nord 2016, I built a web application for the analysis and visualization of the distribution of family names.

I used (among other tools) Node.js, Express, Mocha, AngularJS, D3.js, Java 8, Neo4J, MariaDB, and Docker.

Slides for the lightning talk at Coding da Vinci Nord 2016 (German):

Some screenshots:

The application supports the following use cases:

• Data
  • Viewing the raw data
  • Documenting the enrichment data used for maps and postal codes
  • Known data issues
• Analysis: searching for a name in the dataset
  • Exact match
  • LIKE-pattern search
  • Regular-expression search
  • Finding similar names using Levenshtein distance
• Visualization
  • Geographically on a map of Germany
  • Over time on a timeline
  • Name similarity as a network/graph using Levenshtein distance

The application architecture is shown in the following diagram:

Technologies used:

• Preprocessing: Java 8 with parallel streams
• Databases: Neo4J, MariaDB
• Server: Node.js, Express, Mocha, Chai
• Client: AngularJS, D3.js, Bootstrap

More information:

• Source code on GitHub.
• Project page on HackDash.